https://sven-li-sankyuu.github.io/tags/federal-reserve/Federal ReserveHugo Blox Builder (https://hugoblox.com)en-usTue, 12 Aug 2025 00:00:00 +0000https://sven-li-sankyuu.github.io/media/icon_hu15255132113151062080.pnghttps://sven-li-sankyuu.github.io/tags/federal-reserve/https://sven-li-sankyuu.github.io/publication/fedspeak-confidence/Tue, 12 Aug 2025 00:00:00 +0000https://sven-li-sankyuu.github.io/publication/fedspeak-confidence/<h2 id="abstract">Abstract</h2> <p>This paper proposes an LLM-based uncertainty-aware framework for interpreting Federal Reserve communications (Fedspeak) and classifying monetary policy stance. The framework incorporates domain-specific reasoning grounded in monetary policy transmission mechanisms and introduces dynamic uncertainty decoding to assess prediction confidence.</p> <h2 id="methodology">Methodology</h2> <ul> <li><strong>Domain Knowledge Integration</strong>: Incorporates monetary policy transmission mechanism knowledge</li> <li><strong>Uncertainty Quantification</strong>: Decomposes perceptual uncertainty into cognitive risk and environmental ambiguity</li> <li><strong>Dynamic Decoding</strong>: Adaptively selects decoding strategies based on model confidence levels</li> </ul> <h2 id="results">Results</h2> <p>The framework achieves competitive performance on policy stance analysis tasks, with uncertainty measures providing reliability indicators for predictions.</p>